JP2021017649A - Machine learning method, machine learning device, machine learning program, communication method, and film deposition device - Google Patents

Abstract

To facilitate decision of an appropriate film deposition condition to a cutting tool.SOLUTION: A state variable including at least one physical amount relating to performance evaluation to a cutting tool and at least one film deposition condition is observed. Reward for a determination result of at least one film deposition condition is calculated on the basis of the state variable. A function for determining at least one film deposition condition from the state variable is updated on the basis of the reward. At least one film deposition condition for obtaining highest reward is determined by repetition of update of the function. At least one film deposition condition is at least one parameter of a vacuum evacuation system, a heating-cooling system, an evaporation source system, a table system and a process gas system. At least one physical amount is at least one of a processable number, a blade edge abrasion evaluation value, a blade edge damage evaluation value, a cutting resistance evaluation value, a finished surface evaluation value and a chip condition observation value.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a technique for learning film formation conditions by machine learning.

In recent years, in order to manufacture a cutting tool with high wear resistance, a hard film such as TiN, TiAlN, CrN has been formed on a base material to be a cutting tool by a physical vapor deposition method (PVD method). (For example, Patent Document 1). In order to manufacture a tool with high wear resistance, it is required to appropriately determine the film forming conditions.

Japanese Unexamined Patent Publication No. 2014-114507

However, conventionally, the film forming conditions have been determined by a skilled engineer based on many years of experience. Further, what the tool maker actually requires for the evaluation of the cutting tool is not the evaluation of the film itself but the performance evaluation of the cutting tool. However, conventionally, the film formation conditions have not been determined in consideration of the performance evaluation of the cutting tool. Therefore, it has been difficult to easily determine the appropriate film forming conditions for the cutting tool.

The present invention has been made to solve such a problem, and is to provide a machine learning method or the like that can easily determine an appropriate film forming condition for a cutting tool.

In recent years, various services related to machine learning such as deep learning have been provided on the cloud, and it has become possible for users to easily use these services. Therefore, the present inventor conceived the present invention with the finding that an appropriate film forming condition for a cutting tool can be easily determined by machine learning the film forming conditions and the physical quantity related to the performance evaluation of the cutting tool. It came to.

The machine learning method according to one aspect of the present invention is a machine learning method for the machine learning device to determine the film forming conditions of the film forming apparatus for forming the workpiece which is the base material of the cutting tool. The equipment includes a vacuum exhaust system for vacuuming the chamber, a heating and cooling system for heating and cooling the chamber, an evaporation source system for evaporating the target, a table system on which the work is placed, and a process gas in the chamber. A state variable including at least one physical quantity relating to performance evaluation of the cutting tool and at least one film forming condition including a process gas system and an etching system for introducing the above state variables is acquired, and the state variables are obtained based on the state variables. By calculating the reward for the determination result of at least one film forming condition, updating the function for determining the at least one film forming condition from the state variable based on the reward, and repeating the update of the function. The at least one film forming condition for which the reward is most obtained is determined, and the at least one film forming condition is a first parameter for the vacuum exhaust system, a second parameter for the heating / cooling system, and the evaporation. It is at least one of a third parameter related to the source system, a fourth parameter related to the table system, and a fifth parameter related to the process gas system, and the at least one physical quantity is a workable number and a cutting edge wear evaluation value. , At least one of a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip condition observation value.

According to this configuration, there are a first parameter for the vacuum exhaust system, a second parameter for the heating and cooling system, a third parameter for the evaporation source system, a fourth parameter for the table system, and a fifth parameter for the process gas system. Of the workable number, cutting edge wear evaluation value, cutting edge damage evaluation value, cutting resistance measurement value, finished surface evaluation value, and chip condition observation value related to at least one of these film formation conditions and performance evaluation of the cutting tool. At least one physical quantity is observed as a state parameter. Then, the reward for the determination result of the film formation condition is calculated based on the observed state variable, and the function for determining the film formation condition from the state variable is updated based on the calculated reward, and this update is performed. The film formation conditions that are repeated and the most rewards are obtained are learned. Further, in this configuration, machine learning is performed using a physical quantity peculiar to a cutting tool such as a cutting tool wear evaluation value as a physical quantity related to evaluation of film formation conditions. Therefore, in this configuration, appropriate film forming conditions for the cutting tool can be easily determined.

In the above configuration, the first parameter may be at least one of the exhaust velocity, the ultimate pressure, the residual gas type, the residual gas partial pressure, and the PQ characteristic.

According to this configuration, at least one of the exhaust speed, the ultimate pressure, the residual gas type, the residual gas partial pressure, and the PQ characteristic is set as the film forming condition for the vacuum exhaust system, and the machine learning is performed. Appropriate film formation conditions can be determined in consideration of the condition of the exhaust system.

In the above configuration, the second parameter includes the heater temperature of the heater constituting the heating / cooling system, the work temperature which is the temperature of the work, the heating rate of the heater, the heating rate of the work, and the output of the heater. It may be at least one of the temperature accuracy of the heater, the temperature accuracy of the work, the response characteristics of the heater temperature and the work temperature, the temperature distribution of the heater, and the temperature distribution of the work.

According to this configuration, the heater temperature, the work temperature, the heater temperature rise rate, the work temperature rise rate, the heater output, the heater temperature accuracy, the work temperature accuracy, the heater temperature response characteristic, the work temperature response characteristic, Since at least one of the temperature distribution of the heater and the temperature distribution of the work is the film forming condition related to the heating / cooling system and machine learning is performed, the appropriate film forming condition should be taken into consideration in consideration of the state of the heating / cooling system. Can be decided.

In the above configuration, the third parameter is at least one of the target composition, the target thickness, the target manufacturing method, the arc discharge voltage, the arc discharge current, the evaporation source magnetic field, the evaporation source coil current, and the arc ignition characteristic. It may be one.

According to this configuration, at least one of the target composition, target thickness, target fabrication method, arc discharge voltage, arc discharge current, evaporation source magnetic field, evaporation source coil current, and arc ignition characteristics is deposited on the evaporation source system. Since machine learning is performed as a condition, an appropriate film forming condition can be determined in consideration of the state of the evaporation source system.

In the above configuration, the fourth parameter includes a bias voltage for the work, a bias current for the work, the number of abnormal discharges, a time change of the abnormal discharge, a waveform of the bias voltage, a waveform of the bias current, and a rotation speed of the work. , The shape of the work, the amount of the work mounted, the method of mounting the work, and at least one of the materials of the work may be included.

According to this configuration, bias voltage, bias current, number of abnormal discharges, time change of abnormal discharge, waveform of bias voltage, waveform of bias current, number of rotations of work, shape of work, mounting amount of work, mounting method of work Since at least one of the materials of the work is subjected to machine learning as a film forming condition related to the table system, an appropriate film forming condition can be determined in consideration of the state of the table system.

In the above configuration, the fifth parameter may be at least one of the flow rate of the process gas, the type of the process gas, and the pressure of the process gas.

According to this configuration, since at least one of the process gas flow rate, the process gas type, and the process gas pressure is machine-learned as a film forming condition for the process gas system, the state of the process gas system is taken into consideration. Appropriate film formation conditions can be determined.

In the above configuration, the at least one film forming condition may further include a sixth parameter relating to the etching system.

According to this configuration, since machine learning is performed in consideration of the film forming conditions related to the etching system, it is possible to determine an appropriate film forming condition in consideration of the state of the etching system.

In the above configuration, the sixth parameter is a heating current for heating the filament of the etching system, a heating voltage for heating the filament, a diameter of the filament, a discharge current of the filament, and a discharge voltage of the filament. It may be at least one of.

According to this configuration, at least one of the filament heating current, the filament heating voltage, the filament diameter, the filament discharge current, and the filament discharge voltage is machine-learned as a film forming condition for the etching system. Appropriate film formation conditions can be determined in consideration of the state of the etching system.

Each process of the machine learning method described above may be implemented by a machine learning device, or may be implemented in a machine learning program and distributed. This machine learning device may be configured by a server or a film forming apparatus.

The communication method according to another aspect of the present invention is the communication method of the film forming apparatus when machine learning the film forming conditions of the film forming apparatus for forming the workpiece which is the base material of the cutting tool. The equipment includes a vacuum exhaust system for vacuuming the chamber, a heating and cooling system for heating and cooling the chamber, an evaporation source system for evaporating the target, a table system on which the work is placed, and a process gas in the chamber. A state variable including at least one physical quantity related to performance evaluation of the cutting tool and at least one film forming condition including a process gas system, an etching system, and a communication unit for introducing the above state variable is observed. It is transmitted over a network to receive at least one machine-learned film formation condition, the at least one film formation condition being the first parameter for the vacuum exhaust system, the second parameter for the heating and cooling system, and the said. At least one of a third parameter related to the evaporation source system, a fourth parameter related to the table system, and a fifth parameter related to the process gas system, and the at least one physical quantity is a workable number and a cutting edge wear evaluation. It is at least one of a value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip condition observation value.

According to this configuration, information necessary for machine learning of film formation information is provided. Such a communication method can also be implemented in a film forming apparatus.

According to the present invention, appropriate film forming conditions for a cutting tool can be easily determined.

It is an overall block diagram of the film forming apparatus applied to the machine learning system which concerns on embodiment. It is the whole block diagram of the machine learning system in embodiment. It is a flowchart which shows an example of the processing in the machine learning system shown in FIG. It is a figure which shows an example of the film forming condition. It is a figure which shows an example of a physical quantity. It is an overall block diagram of the machine learning system which concerns on the modification of this invention.

FIG. 1 is an overall configuration diagram of a film forming apparatus applied to the machine learning system according to the embodiment. The film forming apparatus 30 is an apparatus for forming a hard film on a work (object to be coated) which is a base material of a cutting tool by an arc ion plating method. The arc ion plating method is a kind of ion plating method in which a solid material is evaporated by using a vacuum arc discharge. The arc ion plating method is suitable for forming a film of a cutting tool because it has a high ionization rate of the evaporated material and can form a film having excellent adhesion. The hard film is, for example, TiN, TiAlN, TiCN, CrN and the like.

The film forming apparatus 30 includes a vacuum exhaust system 510, a heating / cooling system 520, an evaporation source system 530, a table system 540, a process gas system 550, an etching system 560, and a chamber 570.

The vacuum exhaust system 510 includes an exhaust device 511 and evacuates the inside of the chamber 570. The exhaust device 511 includes a pump for exhausting the air in the chamber 570 and the like.

The heating / cooling system 520 includes a heater power supply unit 521 and a heater 522 to heat the work 545. The heater power supply unit 521 is a power supply circuit that supplies electric power to the heater 522. The heater 522 is provided in the chamber 570 and generates heat by the electric power supplied from the heater power supply unit 521. Further, the heating / cooling system 520 cools the work 545 by stopping the heat generation of the heater 522.

The evaporation source system 530 is a system for evaporating a target (film forming material). The evaporation source system 530 includes an arc cathode 531 and an arc power supply unit 532. The arc power supply unit 532 is a power supply circuit that supplies a discharge current to the arc cathode 531. The arc cathode 531 includes a target, and a vacuum arc discharge is generated between the arc cathode 531 and the inner wall of the chamber 570 by the electric power supplied from the arc power supply unit 532. When the vacuum arc discharge is started, a molten region called an arc spot having a diameter of several μm is generated on the surface of the cathode. A high-density current is concentrated on the arc spot, and the cathode surface is instantly melted and evaporated. The surface of the work 545 is formed by this vacuum arc discharge.

In the example of FIG. 1, two pairs of arc cathode 531 and arc power supply unit 532 are shown, but this is an example, and the arc cathode 531 and arc power supply unit 532 may be a pair, or 3 It may be more than one pair.

The table system 540 is a rotary table on which the work 545 is mounted. The table system 540 includes a table 541, a table drive unit 542, and a bias power supply unit 543. The table 541 is provided in the chamber 570. The work 545 is placed on the table 541. The table drive unit 542 includes a motor and the like to rotate the table 541. The bias power supply unit 543 applies a negative potential to the work 545 via the table 541.

The process gas system 550 introduces a process gas for forming a reactive film in the chamber 570.

The etching system 560 includes a filament (not shown) provided between a discharge power supply unit 561, a pair of filament electrodes 562, and a pair of filament electrodes 562. The discharge power supply unit 561 is a power supply circuit that supplies a discharge current to the filament via a pair of filament electrodes 562. The etching system 560 generates argon plasma between the arc cathode 531 and the filament and between the inner wall and the filament of the chamber 570. The surface of the work 545 is cleaned by the generation of this argon plasma. In this cleaning, the inner walls of the arc cathode 531 and chamber 570 function as anodes and the filaments function as cathodes.

The chamber 570 is a container for accommodating the work 545. The inside of the chamber 570 is evacuated by the vacuum exhaust system 510 to maintain the vacuum state.

FIG. 2 is an overall configuration diagram of the machine learning system according to the embodiment. The machine learning system includes a server 10, a communication device 20, and a film forming device 30. The server 10 and the communication device 20 are connected to each other so as to be able to communicate with each other via the network 40. The communication device 20 and the film forming device 30 are connected to each other so as to be able to communicate with each other via the network 50. The network 40 is a wide area communication network such as the Internet. The network 50 is, for example, a local area network. The server 10 is, for example, a cloud server composed of one or more computers. The communication device 20 is, for example, a computer owned by a user who uses the film forming apparatus 30. The communication device 20 functions as a gateway that connects the film forming device 30 to the network 40. The communication device 20 is realized by installing dedicated application software on a computer owned by the user himself / herself. Alternatively, the communication device 20 may be a dedicated device provided to the user by the manufacturer of the film forming apparatus 30. The film forming apparatus 30 is the film forming apparatus described with reference to FIG.

Hereinafter, the configuration of each device will be specifically described. The server 10 includes a processor 100 and a communication unit 101. The processor 100 is a control device including a CPU and the like. The processor 100 includes a reward calculation unit 110, an update unit 120, a determination unit 130, and a learning control unit 140. Each block included in the processor 100 may be realized by the processor 100 executing a machine learning program that causes the computer to function as a server 10 in the machine learning system, or may be realized by a dedicated electric circuit.

The reward calculation unit 110 calculates a reward for the determination result of at least one film forming condition based on the state variable observed by the state observation unit 321.

The update unit 120 updates a function for determining at least one film formation condition from the state variables observed by the state observation unit 321 based on the reward calculated by the reward calculation unit 110. As the function, the action value function described later is adopted.

The determination unit 130 determines at least one film forming condition that gives the most reward by repeating the update of the function.

The learning control unit 140 controls the overall control of machine learning. The machine learning system of the present embodiment learns the film formation conditions by reinforcement learning. Reinforcement learning means that an agent (behavior) selects a certain action based on the environmental situation, changes the environment based on the selected action, and gives the agent a reward for the environmental change, thereby making the agent better. It is a machine learning method that learns the selection of. Q-learning and TD learning can be adopted as reinforcement learning. In the following description, Q-learning will be described as an example. In the present embodiment, the reward calculation unit 110, the update unit 120, the determination unit 130, the learning control unit 140, and the state observation unit 321 described later correspond to the agent. In the present embodiment, the communication unit 101 is an example of a state acquisition unit that acquires a state variable.

The communication unit 101 is composed of a communication circuit that connects the server 10 to the network 40. The communication unit 101 receives the state variable observed by the state observation unit 321 via the communication device 20. The communication unit 101 transmits the film forming conditions determined by the determination unit 130 to the film forming apparatus 30 via the communication apparatus 20.

The communication device 20 includes a transmitter 201 and a receiver 202. The transmitter 201 transmits the state variable transmitted from the film forming apparatus 30 to the server 10, and also transmits the film forming conditions transmitted from the server 10 to the film forming apparatus 30. The receiver 202 receives the state variable transmitted from the film forming apparatus 30, and also receives the film forming conditions transmitted from the server 10.

In addition to the configuration shown in FIG. 1, the film forming apparatus 30 includes a communication unit 310, a processor 320, a memory 330, a sensor unit 340, and an input unit 350.

The communication unit 310 is a communication circuit for connecting the film forming apparatus 30 to the network 50. The communication unit 310 transmits the state variable observed by the state observation unit 321 to the server 10. The communication unit 310 receives the film formation conditions determined by the determination unit 130 of the server 10. The communication unit 310 receives a film formation execution command, which will be described later, determined by the learning control unit 140.

The processor 320 is a control device including a CPU and the like. The processor 320 includes a state observation unit 321, a film formation execution unit 322, and an input determination unit 323. The communication unit 310 transmits the state variable acquired by the state observation unit 321 to the server 10. Each block included in the processor 320 is realized, for example, by executing a machine learning program in which the CPU functions as a film forming apparatus 30 of the machine learning system.

The state observation unit 321 acquires the physical quantity detected by the sensor unit 340 after the film formation is executed. After the film formation is executed, the state observation unit 321 observes a state variable including at least one physical quantity related to the performance evaluation of the cutting tool and at least one film formation condition. Specifically, the state observation unit 321 acquires the film formation conditions based on the measured values of the sensor unit 340. Further, the state observation unit 321 acquires a physical quantity based on the measured value of the sensor unit 340 and the like.

FIG. 4 is a diagram showing an example of film forming conditions. The film formation conditions are roughly classified into the middle category. In the middle category, the first parameter for the vacuum exhaust system 510, the second parameter for the heating / cooling system 520, the third parameter for the evaporation source system 530, the fourth parameter for the table system 540, and the second parameter for the process gas system 550. At least one of the five parameters is included. In addition, the middle classification may include a sixth parameter for the etching system 560.

The first parameter includes at least one of exhaust velocity, ultimate pressure, residual gas type, residual gas partial pressure, and PQ characteristics. The exhaust speed is the speed at which the vacuum exhaust system 510 exhausts the air, the residual gas, and the introduced process gas in the chamber 570. The exhaust speed is calculated, for example, from the performance values of the pumps constituting the vacuum exhaust system 510. Alternatively, the exhaust speed may be a measured value of the flow rate sensor. The ultimate pressure is the pressure in the chamber 570 before the start of the film forming process. The ultimate pressure is calculated from, for example, the performance value of the pumps constituting the vacuum exhaust system 510. Alternatively, the ultimate pressure may be a measured value of the pressure sensor. The residual gas type is a gas remaining in the chamber 570 and is an impurity. Residual gas species are, for example, nitrogen, oxygen, moisture, and hydrogen. The residual gas type is determined based on the partial pressure of the residual gas described later. The residual gas partial pressure is a partial pressure of a plurality of residual gases remaining in the chamber 570. The residual gas partial pressure is obtained by measurement of a vacuum residual gas monitor such as a quadrupole mass spectrometer. The PQ characteristic is a characteristic showing the relationship between the pressure (P) and the flow rate (Q) in the chamber. The PQ characteristic is obtained by calculation from, for example, the flow rate of gas in the chamber 570 detected by the flow rate sensor and the measured value of the pressure sensor.

The second parameters are heater temperature, work temperature, heater temperature rise rate, work temperature rise rate, heater output, heater temperature accuracy, work temperature accuracy, heater temperature / work temperature, heater temperature distribution, work temperature distribution, cooling gas type, Includes at least one of cooling gas pressure and workpiece cooling rate.

The heater temperature is the temperature of the heater 522. The heater temperature is, for example, a measured value of a temperature sensor (thermocouple). The work temperature is the temperature of the work 545. The work temperature is, for example, a measured value of a temperature sensor provided in the vicinity of the work 545. The heater temperature rise rate is the rate of change of the heater temperature when the heater 522 heats up. The heater heating rate is obtained from the time series change of the heater temperature. The work temperature rise temperature is the rate of change of the work temperature when the work 545 raises the temperature. The work heating rate is obtained from the time-series change of the work temperature.

The heater output is the output of the heater 522. The heater output is calculated from the set value of the heater power supply unit 521. The heater output may be calculated from the sensor-measured values of the current value and the voltage value supplied to the heater.

The heater temperature accuracy is a value indicating a variation in the heater temperature. The heater temperature accuracy is calculated from past heater temperature measurements. The work temperature accuracy is a value indicating a variation in the work temperature. Work temperature accuracy is calculated from past work temperature measurements. The heater temperature / work temperature is a response characteristic of the heater 522 to the work 545.

The heater temperature distribution is the temperature distribution of the heater 522. The heater temperature distribution is obtained from the measured values of a plurality of temperature sensors provided around the heater 522. The work temperature distribution is the temperature distribution of the work 545. The work temperature distribution is obtained from the measured values of a plurality of temperature sensors provided around the work 545.

The cooling gas type is information indicating the type of gas for cooling the inside of the chamber 570, and is an input value input in advance. The cooling gas pressure is the pressure of the cooling gas. The cooling gas pressure is a value measured by a pressure sensor provided in the chamber 570. The work cooling rate is the cooling rate of the work 545. The work cooling rate is obtained from the time-series change of the work temperature detected by the temperature sensor provided in the vicinity of the work 545.

The third parameter includes at least one of target composition, target thickness, target manufacturing process, arc discharge voltage, arc discharge current, evaporation source magnetic field, evaporation source coil current, and arc ignition characteristics. The target composition is the composition of the substance constituting the target. The target thickness is the thickness of the target. The target manufacturing method is a method for manufacturing a target. The target composition, target thickness, and target manufacturing method are input values input in advance.

The arc discharge voltage is a voltage supplied by the arc power supply unit 532 to the arc cathode 531 and is a value measured by a sensor. The arc discharge current is a current supplied by the arc power supply unit 532 to the arc cathode 531 and is a value measured by the sensor.

The evaporation source magnetic field is the position and intensity of the magnetic field emitted by the permanent magnetic flux contained in the evaporation source system 530. The evaporation source magnetic field is a pre-input value. The evaporation source coil current is a current flowing through the coil included in the evaporation source system 530 and is a value measured by a sensor. The arc ignition characteristic is the behavior of the voltage and current on the arc surface at the time of arc ignition. The arc ignition characteristic is obtained from the measured values of the arc discharge voltage and the arc discharge current at a certain timing.

The fourth parameter includes at least one of bias voltage, bias current, number of OLs, change in OL time, bias voltage waveform, bias current waveform, work rotation speed, work shape, work mounting amount, work mounting method, and work material. ..

The bias voltage is a bias voltage supplied to the work 545 by the bias power supply unit 543, and is a value measured by the sensor. The bias current is a bias current supplied to the work 545 by the bias power supply unit 543, and is a value measured by the sensor.

The number of times of OL (OverLoad) is the number of abnormal discharges in the table system or the work, and is a value measured by the sensor. The OL time change is the number of OLs per unit time. The bias voltage waveform is a waveform of the bias voltage and is obtained from a value measured by the sensor. The bias voltage waveform is a voltage waveform especially at the time of pulse bias. The bias current waveform is a waveform of the bias current and is obtained from a value measured by the sensor. The work rotation speed is the rotation speed per unit time of the work 545, and includes the rotation speed per unit time of the table 541 and the rotation speed per unit time when the work 545 rotates on the table 541. The work rotation speed is, for example, a value detected by a sensor. The work shape is a numerical value indicating the shape of the work 545, and is an input value input in advance. The work loading amount is the loading amount (for example, weight) of the work 545, and is an input value input in advance. The work mounting method is a method of mounting the work 545 on the table 541, and is an input value input in advance. The work material is the material of the work 545, and is an input value input in advance.

The fifth parameter includes at least one of gas flow rate, gas type, and gas pressure. The gas flow rate is the flow rate of the process gas. The gas type is information indicating the type of process gas. The gas pressure is the pressure of the process gas. These are, for example, sensor detection values.

The sixth parameter includes at least one of filament heating current, filament heating voltage, filament diameter, discharge current, and discharge voltage. The filament heating current is a heating current for heating a pair of filament electrodes 562 constituting the etching system 560, and is a value measured by a sensor. The filament heating voltage is a heating voltage for heating the pair of filament electrodes 562, and is a value measured by a sensor.

The filament diameter is the diameter of each of the pair of filament electrodes 562, and is an input value input in advance. The filament diameter may be calculated by calculation. The discharge current is the discharge current of the pair of filament electrodes 562, and is a value measured by the sensor. The discharge voltage is the discharge voltage of the pair of filament electrodes 562, and is a value measured by the sensor.

FIG. 5 is a diagram showing an example of a physical quantity. Physical quantities are roughly classified into the middle classification. The middle classification includes at least one of a workable number, a cutting edge wear evaluation value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip condition observation value. The processable number is a value indicating how many times the film was peeled off. The number that can be processed is calculated from the usage history when the formed cutting tool is actually used.

The cutting edge wear evaluation value is a value for evaluating the degree of wear of the cutting edge of a cutting tool. The cutting edge wear evaluation value includes at least one of the flank wear amount and the rake face wear amount.

The cutting edge damage evaluation value is a value for evaluating the degree of damage to the cutting edge of a cutting tool. The cutting edge damage evaluation value includes at least one of boundary wear, chipping, breakage, chipping, flaking, welding, plastic deformation, and thermal cracking.

The various parameters constituting the cutting edge wear evaluation value and the cutting edge damage evaluation value may be, for example, measured values obtained by a person directly observing the cutting tool, or the person observes the cutting tool using a microscope. It may be a measured value obtained by the above.

The cutting resistance measurement value is a value indicating the cutting resistance of the cutting tool. Cutting resistance measurements include cutting force and feed force. The finished surface evaluation value is a value for evaluating the finished surface cut by the cutting tool. The chip state observation value is a value indicating the state of chips generated by cutting with a cutting tool. The chip state observation value includes the chip shape and the chip color. The various parameters that make up the cutting resistance measurement value, finished surface evaluation value, and chip condition observation value are measurements obtained by measuring, for example, the force, power, sound, and vibration of the cutting tool during cutting with a sensor. It may be a value.

The reference is returned to FIG. The film forming execution unit 322 controls the film forming operation of the film forming apparatus 30. The input determination unit 323 automatically or manually determines whether or not the process is mass production. When the input determination unit 323 automatically determines whether or not it is a mass production process, when the number of times the condition number input to the input unit 350 exceeds the reference number, the film forming apparatus 30 is in the mass production process. judge. The condition number is an identification number for specifying a certain film forming condition. The film forming condition specified by the condition number includes at least the film forming condition described as Input among the film forming conditions shown in FIG.

When the input determination unit 323 manually determines whether or not it is a mass production process, if data indicating that it is a mass production process is input to the input unit 350, the film forming apparatus 30 determines that it is in the mass production process. .. When in the mass production process, the film forming apparatus 30 does not perform machine learning.

The memory 330 is, for example, a non-volatile storage device, and stores the finally determined optimum film forming conditions and the like. The sensor unit 340 is various sensors used for measuring the film forming conditions exemplified in FIG. 4 and the physical quantity exemplified in FIG. The input unit 350 is an input device such as a keyboard and a mouse.

FIG. 3 is a flowchart showing an example of processing in the machine learning system shown in FIG. In step S1, the learning control unit 140 acquires the input value of the film forming condition input by the user using the input unit 350. The input value acquired here is an input value for the film forming condition described as Input among the film forming conditions listed in FIG.

In step S2, the learning control unit 140 determines at least one film forming condition and a set value for the film forming condition. Here, the film forming condition to be set is at least one of the film forming conditions listed in FIG. 4 other than the film forming condition described as Input, and the set value can be set. It is a film formation condition. Here, the set value of the film formation condition determined corresponds to the action in reinforcement learning.

Specifically, the learning control unit 140 randomly selects set values for each of the film forming conditions to be set. Here, the set value is randomly selected from within a predetermined range for each of the film forming conditions.

In step S3, the learning control unit 140 causes the film forming apparatus 30 to start the film forming operation by transmitting the film forming execution command to the film forming apparatus 30. When the film formation execution command is received by the communication unit 310, the film formation execution unit 322 sets the film formation conditions according to the film formation execution command and starts the film formation operation. The film formation execution command includes an input value of the film formation condition set in step S1 and a set value of the film formation condition determined in step S2.

When the film forming operation is completed, the state observation unit 321 observes the state variable (step S4). Specifically, the state observing unit 321 determines the physical quantity related to the performance evaluation of the cutting tool shown in FIG. 5 and the film forming condition in which the state is observed by a sensor or the like among the film forming conditions shown in FIG. Get as a state variable. The physical quantity may be input to the film forming apparatus 30 by the user operating the input unit 350, for example, or is input to the film forming apparatus 30 by communication between the measuring instrument for measuring the physical quantity and the film forming apparatus 30. May be good. The state observation unit 321 transmits the acquired state variable to the server 10 via the communication unit 310.

In step S5, the determination unit 130 evaluates the physical quantity. Here, the determination unit 130 determines whether or not the physical quantity to be evaluated (hereinafter, referred to as a target physical quantity) among the physical quantities acquired in step S4 has reached a predetermined reference value, thereby determining the physical quantity. evaluate. The target physical quantity is one or more of the physical quantities listed in FIG. When there are a plurality of target physical quantities, there are a plurality of reference values corresponding to each target physical quantity. As the reference value, for example, a predetermined value indicating that the performance of the cutting tool has reached a certain standard can be adopted.

When the determination unit 130 determines that the target physical quantity has reached the reference value (YES in step S6), the determination unit 130 outputs the film formation condition set in step S2 as the final film formation condition (step S7). On the other hand, when the determination unit 130 determines that the physical quantity has not reached the reference value (NO in step S6), the process proceeds to step S8. When there are a plurality of target physical quantities, the determination unit 130 may determine YES in step S6 when all the target physical quantities reach the reference values.

In step S8, the reward calculation unit 110 determines whether or not the target physical quantity is close to the reference value. When the target physical quantity approaches the reference value (YES in step S8), the reward calculation unit 110 increases the reward for the agent (step S9). On the other hand, when the target physical quantity does not approach the reference value (NO in step S8), the reward calculation unit 110 reduces the reward to the agent (step S10). In this case, the reward calculation unit 110 may increase or decrease the reward according to a predetermined increase / decrease value of the reward. When there are a plurality of target physical quantities, the reward calculation unit 110 may perform the determination in step S8 for each of the plurality of target physical quantities. In this case, the reward calculation unit 110 may increase or decrease the reward for each of the plurality of target physical quantities based on the determination result in step S8. Further, the increase / decrease value of the reward may be different depending on the target physical quantity.

In step S11, the update unit 120 updates the action value function using the reward given to the agent. The Q-learning adopted in the present embodiment is a method of learning a Q-value (Q (s, a)) which is a value for selecting an action a under a certain environmental state s. Incidentally, environmental conditions s _t corresponds to the state variable of the flow. Then, in Q-learning, the action a having the highest Q (s, a) is selected in a certain environmental state s. In Q-learning, various actions a are taken under a certain environmental state s by trial and error, and the correct Q (s, a) is learned using the reward at that time. Action-value function _{Q (s t, a} t) update equation of is expressed by the following equation (1).

_Here, s t, _{a t} each represent an action and environmental condition at time t. Behavioral _{a t,} the environmental conditions are changed to _{s t + 1,} by a change in the environmental conditions, reward _{r t + 1} is calculated. In addition, the term with max is the Q value (Q ( _{st + 1} , a)) when the most valuable action a known at that time is selected under the environmental state _{st + 1} , multiplied by γ. Is. Here, γ is a discount rate, and takes a value of 0 <γ ≦ 1 (usually 0.9 to 0.99). α is a learning coefficient and takes a value of 0 <α ≦ 1 (usually about 0.1).

The update equation is the Q value of the action a in state _{s Q (s t, a t} ) than, based on the Q value when taken the best behavior in the following environmental conditions s _{t + 1} by action a gamma · the larger the better maxQ of _{(s t + 1, a)} , Q (s t, a t) to increase the. On the other hand, the update _{equation, Q (s t, a t} ) is smaller is better in gamma · than _{maxQ (s t + 1, a} ), Q (s t, a t) to reduce. That is, the value of the action a with the certain state s _t, it due to approach the value of the best behavior in the next state s _{t + 1.} As a result, the optimum state for forming the work 545, that is, at least one optimum film forming condition is determined.

When the process of step S11 is completed, the process returns to step S2, the set value of the film forming condition is changed, and the action value function is updated in the same manner. The update unit 120 has updated the action value function, but the present invention is not limited to this, and the action value table may be updated.

Conventionally, in a film forming apparatus, the film forming conditions have been developed by changing the film forming conditions so that a good film can be obtained. In order to obtain a good film, it is necessary to find a relationship between the evaluation of the film and the film forming conditions. However, as shown in FIG. 4, since the types of film forming conditions are enormous, an extremely large number of physical models are required to define such relationships, and such relationships are described by the physical models. It was found that it is difficult to do so. Furthermore, in order to construct such a physical model, it is also required to artificially find out which parameter influences the evaluation of which film, and this construction is difficult. According to the present embodiment, at least one of the above-mentioned first to sixth parameters, the number of machines that can be machined, the cutting edge wear evaluation value, the cutting edge damage evaluation value, and the cutting resistance measurement regarding the performance evaluation of the cutting tool. At least one physical quantity of the value, the finished surface evaluation value, and the chip-like observation value is observed as a state variable. Then, the reward for the determination result of the film formation condition is calculated based on the observed state variable, and the action value function for determining the film formation condition from the state variable is updated based on the calculated reward. The renewal is repeated to learn the film formation conditions that give the most reward. As described above, in this embodiment, the film forming conditions are determined by machine learning without using the above-mentioned physical model. As a result, in this embodiment, appropriate film forming conditions for the cutting tool can be easily determined.

The following modifications can be adopted in the present invention.

(1) FIG. 6 is an overall configuration diagram of a machine learning system according to a modified example of the present invention. The machine learning system according to this modification is composed of the film forming apparatus 30A alone. The film forming apparatus 30A includes a processor 320A, an input unit 391, and a sensor unit 392. The processor 320A includes a machine learning unit 370 and a film forming unit 380. The machine learning unit 370 includes a reward calculation unit 371, an update unit 372, a determination unit 373, and a learning control unit 374. The reward calculation unit 371 to the learning control unit 374 are the same as the reward calculation unit 110 to the learning control unit 140 shown in FIG. 2, respectively. The state observation unit 381, the film formation execution unit 382, and the input determination unit 383 are the same as the state observation unit 321, the film formation execution unit 322, and the input determination unit 323 shown in FIG. 2, respectively. The input unit 391 and the sensor unit 392 are the same as the input unit 350 and the sensor unit 340 shown in FIG. 2, respectively. In this modification, the state observation unit 381 is an example of a state acquisition unit that acquires state information.

As described above, according to the machine learning system according to this modification, the optimum film forming conditions can be learned by the film forming apparatus 30A alone.

(2) In the above flow, the state variables were observed after the film formation operation was completed, but this is an example, and a plurality of state variables may be observed during one film formation operation. For example, when the state variables are composed only of parameters that can be measured instantaneously, a plurality of state variables can be observed during one film formation operation. As a result, the learning time can be shortened.

(3) In the above embodiment, the film forming apparatus 30 is an apparatus for forming a film by the arc ion plating method, but the present invention is not limited to this, and is formed by another physical vapor deposition method such as a vapor deposition method. It may be a device for filming.

10: Server 30: Film formation device 100: Processor 110: Reward calculation unit 120: Update unit 130: Decision unit 140: Learning control unit 510: Vacuum exhaust system 520: Heating and cooling system 530: Evaporation source system 540: Table system 550: Process gas system 560: Etching system 570: Chamber 574: Learning control unit

Claims (12)

It is a machine learning method in which the machine learning device determines the film forming conditions of the film forming apparatus for forming the work which is the base material of the cutting tool.
The film forming apparatus includes a vacuum exhaust system for evacuating a chamber, a heating / cooling system for heating and cooling the chamber, an evaporation source system for evaporating a target, a table system on which a work is placed, and the chamber. Including a process gas system that introduces process gas into the system and an etching system,
A state variable including at least one physical quantity related to the performance evaluation of the cutting tool and at least one film forming condition is acquired.
Based on the state variable, the reward for the determination result of at least one film forming condition is calculated.
A function for determining at least one film formation condition from the state variable is updated based on the reward.
By repeating the update of the function, the at least one film forming condition from which the reward is most obtained is determined.
The at least one film forming condition includes a first parameter relating to the vacuum exhaust system, a second parameter relating to the heating and cooling system, a third parameter relating to the evaporation source system, a fourth parameter relating to the table system, and the process. At least one of the fifth parameters for the gas system,
The at least one physical quantity is at least one of a workable number, a cutting edge wear evaluation value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip state observation value.
Machine learning method. The first parameter is at least one of the exhaust rate, the ultimate pressure, the residual gas type, the residual gas partial pressure, and the PQ characteristic.
The machine learning method according to claim 1. The second parameters are the heater temperature of the heater constituting the heating / cooling system, the work temperature which is the temperature of the work, the temperature rise rate of the heater, the temperature rise rate of the work, the output of the heater, and the temperature of the heater. At least one of accuracy, temperature accuracy of the work, heater temperature and response characteristics of the work temperature, temperature distribution of the heater, and temperature distribution of the work.
The machine learning method according to claim 1 or 2. The third parameter is at least one of the target composition, the target thickness, the target manufacturing method, the arc discharge voltage, the arc discharge current, the evaporation source magnetic field, the evaporation source coil current, and the arc ignition characteristic.
The machine learning method according to any one of claims 1 to 3. The fourth parameter includes a bias voltage for the work, a bias current for the work, the number of abnormal discharges, a time change of the abnormal discharge, a waveform of the bias voltage, a waveform of the bias current, a rotation speed of the work, and the number of rotations of the work. Includes at least one of shape, loading amount of the work, mounting method of the work, and material of the work.
The machine learning method according to any one of claims 1 to 4. The fifth parameter is at least one of the flow rate of the process gas, the type of the process gas, and the pressure of the process gas.
The machine learning method according to any one of claims 1 to 5. The at least one film forming condition further includes a sixth parameter relating to the etching system.
The machine learning method according to any one of claims 1 to 6. The sixth parameter is at least one of a heating current for heating the filament of the etching system, a heating voltage for heating the filament, a diameter of the filament, a discharge current of the filament, and a discharge voltage of the filament. Is,
The machine learning method according to claim 7. It is a machine learning device that determines the film formation conditions of the film formation device that deposits the workpiece that is the base material of the cutting tool.
The film forming apparatus includes a vacuum exhaust system for evacuating a chamber, a heating / cooling system for heating and cooling the chamber, an evaporation source system for evaporating a target, a table system on which a work is placed, and the chamber. Including a process gas system that introduces process gas into the system and an etching system,
A state acquisition unit for observing a state variable including at least one physical quantity related to performance evaluation of the cutting tool and at least one film forming condition.
A reward calculation unit that calculates a reward for the determination result of at least one film formation condition based on the state variable.
An update unit that updates a function for determining at least one film formation condition based on the state variable based on the reward.
By repeating the update of the function, the determination unit for determining at least one film forming condition for which the reward is most obtained is provided.
The at least one film forming condition includes a first parameter relating to the vacuum exhaust system, a second parameter relating to the heating and cooling system, a third parameter relating to the evaporation source system, a fourth parameter relating to the table system, and the process. At least one of the fifth parameters for the gas system,
The at least one physical quantity is at least one of a workable number, a cutting edge wear evaluation value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip state observation value.
Machine learning device. A computer-readable machine learning program that allows a computer to function as a machine learning device that determines the film formation conditions of a film forming device that deposits a workpiece that is the base material of a cutting tool.
The film forming apparatus includes a vacuum exhaust system for evacuating a chamber, a heating / cooling system for heating and cooling the chamber, an evaporation source system for evaporating a target, a table system on which a work is placed, and the chamber. Including a process gas system that introduces process gas into the system and an etching system,
A state acquisition unit for observing a state variable including at least one physical quantity related to performance evaluation of the cutting tool and at least one film forming condition.
A reward calculation unit that calculates a reward for the determination result of at least one film formation condition based on the state variable.
An update unit that updates a function for determining at least one film formation condition based on the state variable based on the reward.
By repeating the update of the function, the determination unit for determining at least one film forming condition for which the reward is most obtained is provided.
The at least one film forming condition includes a first parameter relating to the vacuum exhaust system, a second parameter relating to the heating and cooling system, a third parameter relating to the evaporation source system, a fourth parameter relating to the table system, and the process. At least one of the fifth parameters for the gas system,
The at least one physical quantity is at least one of a workable number, a cutting edge wear evaluation value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip state observation value.
Machine learning program. It is a communication method of the film forming apparatus when machine learning the film forming conditions of the film forming apparatus for forming a work which is a base material of a cutting tool.
The film forming apparatus includes a vacuum exhaust system for evacuating a chamber, a heating / cooling system for heating and cooling the chamber, an evaporation source system for evaporating a target, a table system on which a work is placed, and the chamber. Includes a process gas system that introduces process gas into the system, an etching system, and a communication unit.
Observe the state variables including at least one physical quantity related to the performance evaluation of the cutting tool and at least one film forming condition.
The state variable is transmitted over the network to receive at least one machine-learned film formation condition.
The at least one film forming condition includes a first parameter relating to the vacuum exhaust system, a second parameter relating to the heating and cooling system, a third parameter relating to the evaporation source system, a fourth parameter relating to the table system, and the process. At least one of the fifth parameters for the gas system,
The at least one physical quantity is at least one of a workable number, a cutting edge wear evaluation value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip state observation value.
Communication method. A film forming device that forms a film on a workpiece that is the base material of a cutting tool.
A vacuum exhaust system for evacuating the chamber,
A heating and cooling system that heats and cools the chamber,
An evaporation source system that evaporates the target,
A table system on which the work is placed and
A process gas system that introduces process gas into the chamber,
Etching system and
A state observing unit that observes a state variable including at least one physical quantity related to performance evaluation of the cutting tool and at least one film forming condition.
It includes a communication unit that transmits the state variables on the network and receives at least one machine-learned film formation condition.
The at least one film forming condition includes a first parameter relating to the vacuum exhaust system, a second parameter relating to the heating and cooling system, a third parameter relating to the evaporation source system, a fourth parameter relating to the table system, and the process. At least one of the fifth parameters for the gas system,
The at least one physical quantity is at least one of a workable number, a cutting edge wear evaluation value, a cutting edge damage evaluation value, a cutting resistance measurement value, a finished surface evaluation value, and a chip state observation value.
Film forming equipment.

Images